INDEX

Explanations

academic language and terminology related to proofs, comparisons, and summarization of results

oai_token-act-pair · gpt-4o-mini Triggered by @bot

followed by "that" or "of"

np_acts-logits-general · gemini-2.0-flash

summarizing and explaining

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Wikimedijinoj

-0.77

MessageTagHelper

-0.73

⟬

-0.70

StoryboardSegue

-0.64

WebVitals

-0.63

 resourceCulture

-0.63

 gynhyrchwyd

-0.61

StructEnd

-0.60

OGND

-0.60

الإنجليزية

-0.60

POSITIVE LOGITS

ksesta

0.35

shown

0.34

mentioned

0.33

 relâche

0.33

strongly

0.33

ποι

0.31

 Fußballspieler

0.31

+#+#

0.31

 also

0.31

also

0.31

Activations Density 0.443%

academic language and terminology related to proofs, comparisons, and summarization of results

followed by "that" or "of"

summarizing and explaining

No Comments

No Known Activations